Regulatory T cells (Tregs) comprise a heterogeneous class of lymphocytes that are able to promote immune tolerance in peripheral tissue through cytokine secretion and modulation of dendritic cell function. Methods currently exist to selectively expand Tregs in vitro in presence of TGF-b (induced or iTregs) and transfer to patients for therapeutic inhibition of inflammation in those suffering from inflammatory bowel disease, graft- versus-host disease and other pathologies characterized by excessive inflammation. Although the adoptive transfer of iTregs has the promise to be safe in the clinic, major hurdles such as still exist in the translation of this therapeutic strategy. For example, the iTregs transferred into a patient with acute or chronic inflammation could transition from the anti inflammatory, iTreg phenotype to a pro-inflammatory, Th17 phenotype under the influence of inflammatory cytokines present in the microenvironment, and thereby, negatively contribute to the inflammatory state. Therefore, it is important to generate iTregs that possess a stable regulatory phenotype and function when introduced into an inflammatory microenvironment. Several studies link in vivo Treg development to the presence of an abundant and diverse microbial population in the intestinal tract (the microbiota). Although the microbiota's role in physiologic immune tolerance is poorly understood, a prevalent hypothesis is that the microbiota produces specific factors that promote Treg induction and modulate gut immunity towards a tolerant state. We previously demonstrated that indole, a microbiota metabolite derived from dietary tryptophan and present in the GI tract of both healthy mice and humans, attenuate indicators of inflammation. Unpublished data from our lab also show that, after conditioning in vitro in the presence of indole and iTreg-skewing conditions, CD4+ CD25- nave T cells dramatically expand into Foxp3+ iTregs. However, iTreg stability and function can be synergistically increased or attenuated by several pro- and anti-inflammatory cytokines, and therefore, the ability to systematically predict the relationship of microbiota metabolite regulatin of Treg stability and function in inflammatory environments would contribute to our understanding of Treg immunobiology and advance Treg cell-based therapy. Our overall hypothesis is that tryptophan derived microbiota metabolites (TDMMs) induce a Treg phenotype with enhanced stability in vivo. Using Treg and Th17 data from exposure to different tryptophan derived microbiota metabolites, we propose to develop neural network models for Treg induction in vitro and stability post-transfer in vivo in the presence of indole, and use the model for generating testable predictions on optimal Treg induction, function and stability. The specific aims are: (1): To comprehensively determine the phenotype and function of TDMM-induced iTregs and Th17 cells in vitro; (2): Model the effect of TDMM on iTreg induction and Th17 attenuation in vitro and phenotype maintenance in vivo; and (3): To determine the function and stability of TDMM-induced Tregs and TDMM- attenuated Th17 cells after transfer to lymphopenic mice and mice with experimentally induced colitis.